| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 1828352 | Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment | 2009 | 6 Pages |
In this paper a new parametric method to deal with discrepant experimental results is developed. The method is based on the fit of a probability density function to the data. This paper also compares the characteristics of different methods used to deduce recommended values and uncertainties from a discrepant set of experimental data. The methods are applied to the 137Cs and 90Sr published half-lives and special emphasis is given to the deduced confidence intervals. The obtained results are analyzed considering two fundamental properties expected from an experimental result: the probability content of confidence intervals and the statistical consistency between different recommended values. The recommended values and uncertainties for the 137Cs and 90Sr half-lives are 10,984 (24) days and 10,523 (70) days, respectively.
